Setting the bar high or setting up to fail? Interpretations and implications of the EXPLORE Study (HPTN 015)

Abstract

Controlled studies show that HIV risk reduction counseling significantly increases condom use, reduces unprotected sex and prevents sexually transmitted infections. Nevertheless, without evidence of reducing HIV incidence, these interventions are generally discarded. One trial, the EXPLORE study, was designed to test whether ten sessions of risk reduction counseling could impact HIV incidence among men who have sex with men in six US cities. Based on epidemiologic models to define effective HIV vaccines, a 35 % reduction in HIV incidence was set a priori as the benchmark of success in this behavioral intervention trial. Results demonstrated a significant effect of the intervention, with more than a 35 % reduction in HIV incidence observed during the initial 12-18 months following counseling. Over an unprecedented 48-month follow-up, however, the effect of counseling on HIV incidence declined to 18 %. The current review examined how the scientific literature has thus far judged the outcomes of the EXPLORE study as well as the policy implications of these judgments. We identified 127 articles that cited the EXPLORE study since its publication. Among articles that discuss the HIV incidence outcomes, 20 % judged the intervention effective and 80 % judged the intervention ineffective. The overwhelmingly negative interpretation of the EXPLORE study outcomes is reflected in public policies and prevention planning. We conclude that using a vaccine standard to define success led to a broad discrediting of the benefits of behavioral counseling and, ultimately, adversely impacted policies critical to the field of HIV prevention.

Figure 1

Number of Articles Citing Project Respect (1999–2006) and the EXPLORE study (2005–2012).